Shandong Civil Engineering Disaster Prevention and Mitigation Laboratory, Shandong University of Science and Technology, Qingdao, China.
School of Civil Engineering and Architecture, Shandong University of Science and Technology, Qingdao, China.
PLoS One. 2023 Aug 11;18(8):e0288633. doi: 10.1371/journal.pone.0288633. eCollection 2023.
China is gradually transitioning from the "tunnel construction exploration era" to the "tunnel high-quality construction and operation era", and the maintenance demand of highway tunnels has increased sharply. Therefore, there is an urgent need for an evaluation method to evaluate the service reliability of highway tunnels, so as to provide reference for tunnel maintenance personnel to carry out maintenance work. Taking highway tunnels as the research object, this paper extracts three parameters, including length, maximum width and fractal dimension, from the binary image of highway tunnel lining cracks. The standard for dividing the length of the highway tunnel section is 500m as the tunnel section, and a section disease sample space including multiple highway tunnels is constructed. The EM clustering algorithm was used to determine the number of graded grades of disease, and the relative Euclidean distance was used as the evaluation index to divide the safety grade of the tunnel into five grades: normal, degraded, inferior, deteriorated and hazardous. The partial least squares method is used to establish the lining service reliability evaluation formula and verify the residual of each sample point in the sample space. The smaller the average value of the residual, the better fitting effect of the established evaluation formula. The service reliability evaluation method proposed in this paper is applied to engineering practice and compared with the expert scoring method and the national standard method, which proves that the evaluation method in this paper has the advantages of strong visibility, simple evaluation method, and is conducive to engineering practice.
中国正逐步从“隧道建设探索时代”过渡到“隧道高质量建设和运营时代”，公路隧道的维护需求急剧增加。因此，迫切需要一种评估方法来评估公路隧道的服务可靠性，为隧道维护人员开展维护工作提供参考。本文以公路隧道为研究对象，从公路隧道衬砌裂缝的二值图像中提取长度、最大宽度和分形维三个参数。公路隧道段的划分标准以 500m 为一个隧道段，构建一个包含多个公路隧道的节段病害样本空间。利用 EM 聚类算法确定病害分级的级数，并采用相对欧氏距离作为评价指标，将隧道的安全等级分为五个等级：正常、降级、较差、劣化和危险。利用偏最小二乘法建立衬砌服务可靠性评价公式，并对样本空间中各样本点的残差进行验证。样本点残差的平均值越小，建立的评价公式拟合效果越好。将本文提出的服务可靠性评价方法应用于工程实践，并与专家评分法和国家标准法进行比较，验证了本文评价方法具有可视化程度高、评价方法简单、有利于工程实践等优点。
